Validity of the Assessment of Quality of Life (AQoL) utility instrument in patients with operable and inoperable lung cancer
Introduction
Health-related quality of life (HRQoL) is now commonly evaluated as an important outcome measure in lung cancer clinical trials. However, many of these trials focus on patients with more advanced stages of lung cancer, using disease-specific measures [1], [2], [3]. Few studies have incorporated measures of health state preferences or utility [4], [5], [6], [7]. A utility measurement is an index that represents an estimate of the strength of preference for a health outcome/health state relative to other health states [4], [5], [6]. The index can be combined with data on the duration of a particular health state to calculate the number of quality adjusted life years as a basis for a cost-utility analysis [4], [5], [6]. For cost-utility analyses of interventions that might impact on all stages of lung cancer, such as screening, data are needed on the utility associated with both early and late stage disease. After searching the literature, the authors were unable to find any longitudinal studies of HRQoL and utility in patients with all stages of non-small cell lung cancer (NSCLC) at presentation. For example, utility values used in previous cost-utility analyses of lung cancer screening have been based on expert opinion or cross-sectional surveys of utility [6], [7], [8], [9], [10]. Cross-sectional studies do not provide data on changes in HRQoL and utility that might occur over time as a result of treatment or disease progression.
Disease-specific lung cancer questionnaires have been developed and validated predominantly in patients with more advanced lung cancer and often those undergoing chemotherapy or radiotherapy in the clinical trial setting [1], [2], [11], [12]. On the other hand, HRQoL in long-term survivors of lung cancer or those with early stage disease who have undergone surgical resection has usually been evaluated using generic HRQoL instruments [13], [14], [15], [16], [17]. In particular, the Medical Outcomes 36-item Short Form Health Survey (SF-36) has commonly been used and the findings of one study support the construct validity of the SF-36 for patients undergoing surgery for lung cancer [16]. There has been one prospective longitudinal study of HRQoL that examined HRQoL using both a disease-specific and generic quality of life instrument in patients with all lung cancer stages, but this study only included six patients initially treated with surgery [18].
Several different methods may be used for obtaining utility values for different health states [5], [19], [20]. A multi-attribute utility instrument (MAU), however, is efficient to administer and can be used to follow changes in health state profiles over time. Data derived from a generic MAU can be used in economic evaluations that are aimed at examining allocative efficiency across a broad range of interventions or diseases and is therefore appropriate for the public health/screening setting. There are now several generic multi-attribute utility instruments in the published literature, however, it is not clear which of these are most appropriate for use in patients with lung cancer [7], [20], [21], [22], [23], [24], [25]. The Assessment of Quality of Life instrument (AQoL) is a generic MAU that has been calibrated using Australian (general population) preference scores based on the time trade off method [26]. This instrument has been psychometrically validated in other patient groups and used over multiple trials in Australia [26], [27], [28]. The present study was designed to examine the psychometric properties of the AQoL in patients with lung cancer. In particular, the reliability was examined and the concurrent validity was assessed using the SF-36 as the comparator instrument. The sensitivity to different health states of the AQoL and the responsiveness to change over time was also examined.
Section snippets
Methods
A prospective, non-experimental cohort study was undertaken to evaluate HRQoL and utility in patients with lung cancer. This was an observational study and did not involve any clinical protocols.
Health status scores and changes in health status over time
The distribution of baseline AQoL and PCS SF-36 and MCS SF-36 scores was skewed and therefore to describe the data for the study population the median and interquartile range was reported for all participants and according to different clinical and demographic subgroups. For comparisons between means, however, parametric tests were used. In addition, mean values and 95% confidence intervals for the mean were reported for AQoL, PCS and MCS scores so that the results could be viewed in the
Reliability of the AQoL
The internal consistency of the AQoL was assessed using Cronbach's Alpha [37]. Internal consistency is a measure of the homogeneity of an instrument, or the degree to which individual items in the questionnaire reflect a single trait or construct. There is a lack of consensus in the literature about how high the internal consistency should be, however, some experts have recommended that 0.7–0.9 would be an acceptable range [37].
Validity: correlation between AQoL, SF-36, and ECOG performance status
In the psychometric context, the term validity refers to whether an instrument is measuring what it was designed to measure [38]. HRQoL and utility are constructs and there is no gold standard for comparison. The concurrent validity of the AQoL in lung cancer patients was assessed by examining the correlation with another independent measure of HRQoL (the SF-36). In particular, the divergent and convergent validity of the AQoL was assessed [39]. (Concurrent validity is assessed by correlating a
Survival analysis at 6 months in individuals with inoperable NSCLC
The 6-month overall survival was assessed in patients with inoperable NSCLC using the Kaplan–Meier method. Survival was assessed from the date of enrolment in the study. The vital status of all participants was known for the 6-month follow-up point. The influence of variables on survival was analysed using the Cox proportional hazards model in both univariate and multivariate analysis. For the multivariate analysis, factors known to influence survival and AQoL utility scores, SF-36 PCS and MCS
Responsiveness of the AQoL and SF-36
Responsiveness has been defined as the ability of an instrument to measure a clinically important change or meaningful change in a clinical state [41]. The relative efficiency statistic was calculated to compare the responsiveness of the AQoL and its subscales and the SF-36 subscales in patients undergoing lung cancer resection [42]. In particular, the square of the t-statistic (from the paired t-test for the comparison between 3 months and baseline scores) for each subscale, relative to the
Results
The mean age of study participants was 67 (S.D. = 9.3) years and 31.5% were females. The majority had NSCLC. The demographic and clinical characteristics of the study group are listed in Table 1, Table 2, respectively. Early stage lung cancers are over-represented in comparison with epidemiological data and with data from the South Australian hospital-based registry [44], [45], [46]. This may reflect the fact that participants were recruited from an ambulatory multi-disciplinary clinic in a
Discussion
To the authors’ knowledge this is the first longitudinal study of generic HRQoL using a MAU instrument to be conducted in individuals with all stages of NSCLC. The study findings highlight that HRQoL changes over time. At the time of initial diagnosis, the mean utility of patients with lung cancer was 0.62. This compares with a mean utility of 0.79 in a general population sample and with a mean value of 0.63 obtained for a sample of outpatients attending two large Australian public hospitals
Conclusion
The data from this study, to some extent, support the validity of the AQoL for use in patients with lung cancer. However, there remains some uncertainty about whether the AQoL has sufficient content validity and sensitivity to different health states for use in patients with lung cancer. Furthermore, because of the small numbers in the study, some of the estimates of utilities associated with different health states in individuals with the lung cancer are imprecise. Further studies using other
Acknowledgements
We are grateful to all the staff at St. Vincent's hospital who assisted with the recruitment of patients for this study, including Ms. Monica Lammers, Ms. Maria Loder, and other members of the Oncology and Respiratory Units. We also thank Ms. Monica Lammers and Ms. Maria Loder for providing assistance with maintaining the database of participants in this study. We further extend our gratitude to all the patients who took part in this survey. Renee Manser was supported by a NHMRC postgraduate
References (52)
- et al.
Quality of life in patients with lung cancer: a review of literature from 1970 to 1995
Chest
(1998) - et al.
The EORTC QLQ-LC13: a modular supplement to the EORTC Core Quality of Life Questionnaire (QLQ-C30) for use in lung cancer clinical trials
Eur J Cancer
(1994) Utility approach to measuring health-related quality of life
J Chron Dis
(1987)- et al.
Economic decision analysis model of screening for lung cancer
Eur J Cancer
(2001) - et al.
The effect of patient's risk taking attitude on the cost effectiveness of testing strategies in the evaluation of pulmonary lesions
Chest
(1997) - et al.
Reliability and validity of the functional assessment of cancer therapy-lung (FACT-L) quality of life instrument
Lung Cancer
(1995) - et al.
What happens to patients undergoing lung cancer surgery: outcomes and quality of life before and after surgery
Chest
(2002) - et al.
Quality of life in lung cancer patients: an important prognostic factor
Lung Cancer
(2001) Cost utility analysis: what should be measured?
Soc Sci Med
(1994)- et al.
Quality of life assessment in the community-dwelling elderly: validation of the Assessment of Quality of Life (AQoL) Instrument and comparison with the SF-36
J Clin Epidemiol
(2003)